Acute myeloid leukemia (AML) comprises a heterogeneous group of highly lethal, clonal disorders of blood cell progenitors for which more effective therapies are urgently needed. The underlying genetic abnormalities transform hematopoietic progenitors into AML blasts that are resistant to apoptosis and fail to undergo normal differentiation. While conventional chemotherapy regimens frequently ablate actively cycling leukemic blasts and induce remissions, most patients relapse with abysmal outcomes. Recent studies highlight an important role for stem cell transcriptional networks in supporting survival and resistance to chemotherapy in leukemic blasts, although druggable targets that maintain these networks have remained elusive. We previously discovered that the cMYC oncoprotein activates HMGA1 expression in experimental models of lymphoid leukemia, while HMGA1 directly induces cMYC in embryonic stem cells. Here, we demonstrate for the first time that the HMGA1-cMYC network is a therapeutic target that is critical for leukemic cell survival and resistance to chemotherapy.

In primary AML blasts, we found that both HMGA1 and cMYC gene expression are enriched relative to levels in normal CD34+ hematopoietic stem and progenitor cells and their expression is positively correlated. Silencing HMGA1 or cMYC in AML blasts dramatically halts cell growth, leading to rapid apoptotic cell death. HMGA1 transcriptionally activates cMYC in leukemic blasts, while the cMYC oncoprotein, in turn, induces HMGA1, resulting in potent feed-forward loop that maintains survival in AML blasts.We also discovered that AML blasts exposed to AraC (cytosine arabinoside), a standard cytotoxic agent used in AML, become resistant by up-regulating both HMGA1 and cMYC. Moreover, silencing either HMGA1 or cMYC restores sensitivity to AraC, demonstrating that induction of HMGA1 and cMYC is functionally important. Next, we sought to determine whether the inhibitor of the BET family of bromodomain proteins, JQ1, has anti-leukemic effects in AML blasts resistant to AraC because it functions, at least in part, by disrupting cMYC transcriptional complexes. We found that cMYC is repressed within 4 hours, while HMGA1 requires 24 hours for down-regulation following exposure to JQ1. Moreover, JQ1 restores sensitivity to AraC in the resistant AML blasts.

Together, these studies reveal a novel role for HMGA1 and cMYC in a feed-forward loop that is critical for AML blast survival and chemoresistance. These findings also suggest that JQ1 will be effective therapy in AML blasts that are resistant to AraC.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.